The tremendous development in large language models (LLM) has led to a new wave of innovations and applications and yielded research results that were initially forecast to take longer. In this work, we tap into these recent developments and present a meta-study about the potential of large language models if deployed in social robots. We place particular emphasis on the applications of social robots: education, healthcare, and entertainment. Before being deployed in social robots, we also study how these language models could be safely trained to ``understand'' societal norms and issues, such as trust, bias, ethics, cognition, and teamwork. We hope this study provides a resourceful guide to other robotics researchers interested in incorporating language models in their robots.

Operating systems (OSes) are foundational to computer systems, managing hardware resources and ensuring secure environments for diverse applications. However, despite their enduring importance, the fundamental design objectives of OSes have seen minimal evolution over decades. Traditionally prioritizing aspects like speed, memory efficiency, security, and scalability, these objectives often overlook the crucial aspect of intelligence as well as personalized user experience. The lack of intelligence becomes increasingly critical amid technological revolutions, such as the remarkable advancements in machine learning (ML). Today's personal devices, evolving into intimate companions for users, pose unique challenges for traditional OSes like Linux and iOS, especially with the emergence of specialized hardware featuring heterogeneous components. Furthermore, the rise of large language models (LLMs) in ML has introduced transformative capabilities, reshaping user interactions and software development paradigms. While existing literature predominantly focuses on leveraging ML methods for system optimization or accelerating ML workloads, there is a significant gap in addressing personalized user experiences at the OS level. To tackle this challenge, this work proposes PerOS, a personalized OS ingrained with LLM capabilities. PerOS aims to provide tailored user experiences while safeguarding privacy and personal data through declarative interfaces, self-adaptive kernels, and secure data management in a scalable cloud-centric architecture; therein lies the main research question of this work: How can we develop intelligent, secure, and scalable OSes that deliver personalized experiences to thousands of users?

Computational notebooks became indispensable tools for research-related development, offering unprecedented interactivity and flexibility in the development process. However, these benefits come at the cost of reproducibility and an increased potential for bugs. There are many tools for bug fixing; however, they are generally targeted at the classical linear code. With the rise of code-fluent Large Language Models, a new stream of smart bug-fixing tools has emerged. However, the applicability of those tools is still problematic for non-linear computational notebooks. In this paper, we propose a potential solution for resolving errors in computational notebooks via an iterative LLM-based agent. We discuss the questions raised by this approach and share a novel dataset of computational notebooks containing bugs to facilitate the research of the proposed approach.

State-of-the-art intent classification (IC) and slot filling (SF) methods often rely on data-intensive deep learning models, limiting their practicality for industry applications. Large language models on the other hand, particularly instruction-tuned models (Instruct-LLMs), exhibit remarkable zero-shot performance across various natural language tasks. This study evaluates Instruct-LLMs on popular benchmark datasets for IC and SF, emphasizing their capacity to learn from fewer examples. We introduce ILLUMINER, an approach framing IC and SF as language generation tasks for Instruct-LLMs, with a more efficient SF-prompting method compared to prior work. A comprehensive comparison with multiple baselines shows that our approach, using the FLAN-T5 11B model, outperforms the state-of-the-art joint IC+SF method and in-context learning with GPT3.5 (175B), particularly in slot filling by 11.1--32.2 percentage points. Additionally, our in-depth ablation study demonstrates that parameter-efficient fine-tuning requires less than 6% of training data to yield comparable performance with traditional full-weight fine-tuning.

GPT-4, and LLaMA, are increasingly prevalent. Numerous studies have explored effective prompting techniques to harness the power of these LLMs for various research problems. Retrieval, specifically in the legal data domain, poses a challenging task for the direct application of Prompting techniques due to the large number and substantial length of legal articles. This research focuses on maximizing the potential of prompting by placing it as the final phase of the retrieval system, preceded by the support of two phases: BM25 Pre-ranking and BERT-based Re-ranking. Experiments on the COLIEE 2023 dataset demonstrate that integrating prompting techniques on LLMs into the retrieval system significantly improves retrieval accuracy. However, error analysis reveals several existing issues in the retrieval system that still need resolution.

This study introduces a novel approach to online embedding of multi-scale CLIP (Contrastive Language-Image Pre-Training) features into 3D maps. By harnessing CLIP, this methodology surpasses the constraints of conventional vocabulary-limited methods and enables the incorporation of semantic information into the resultant maps. While recent approaches have explored the embedding of multi-modal features in maps, they often impose significant computational costs, lacking practicality for exploring unfamiliar environments in real time. Our approach tackles these challenges by efficiently computing and embedding multi-scale CLIP features, thereby facilitating the exploration of unfamiliar environments through real-time map generation. Moreover, the embedding CLIP features into the resultant maps makes offline retrieval via linguistic queries feasible. In essence, our approach simultaneously achieves real-time object search and mapping of unfamiliar environments. Additionally, we propose a zero-shot object-goal navigation system based on our mapping approach, and we validate its efficacy through object-goal navigation, offline object retrieval, and multi-object-goal navigation in both simulated environments and real robot experiments. The findings demonstrate that our method not only exhibits swifter performance than state-of-the-art mapping methods but also surpasses them in terms of the success rate of object-goal navigation tasks.

Large language models (LLM), such as Google's Minerva and OpenAI's GPT families, are becoming increasingly capable of solving mathematical quantitative reasoning problems. However, they still make unjustified logical and computational errors in their reasoning steps and answers. In this paper, we leverage the fact that if the training corpus of LLMs contained sufficiently many examples of formal mathematics (e.g. in Isabelle, a formal theorem proving environment), they can be prompted to translate i.e. autoformalize informal mathematical statements into formal Isabelle code -- which can be verified automatically for internal consistency. This provides a mechanism to automatically reject solutions whose formalized versions are inconsistent within themselves or with the formalized problem statement. We evaluate our method on GSM8K, MATH and MultiArith datasets and demonstrate that our approach provides a consistently better heuristic than vanilla majority voting -- the previously best method to identify correct answers, by more than 12% on GSM8K. In our experiments it improves results consistently across all datasets and LLM model sizes. The code can be found at https://github.com/jinpz/dtv. Recently, language models (Devlin et al., 2018; Brown et al., 2020; Chowdhery et al., 2022) have advanced significantly in many natural language processing tasks such as machine translation, question answering, summarization, etc. More recent large language models (LLMs) such as Minerva (Lewkowycz et al., 2022), GPT3.5 (OpenAI) and GPT4 (OpenAI, 2023) have also become increasingly capable of solving quantitative reasoning problems, ranging from middle school math word problems (Cobbe et al., 2021) to challenging high school mathematical competition problems (Hendrycks et al., 2021). By training or finetuning the model on high-quality natural language mathematical and scientific text, these LLMs can generate self-contained step-by-step solutions to quantitative reasoning problems without relying on external tools. However, just like human beings, the solutions LLMs generate are prone to simple calculation errors and unjustified logical leaps. Following Wang et al. (2022); Lewkowycz et al. (2022), one can sample many proposed solutions, extract the final answer from each, and select the most common answer. While aggregating answers like this improves performance at the problem level, the most common answer is sometimes wrong. Ideally, we would like a better heuristic to identify the correct answer.

Lexical-syntactic flexibility, in the form of conversion (or zero-derivation) is a hallmark of English morphology. In conversion, a word with one part of speech is placed in a non-prototypical context, where it is coerced to behave as if it had a different part of speech. However, while this process affects a large part of the English lexicon, little work has been done to establish the degree to which language models capture this type of generalization. This paper reports the first study on the behavior of large language models with reference to conversion. We design a task for testing lexical-syntactic flexibility -- the degree to which models can generalize over words in a construction with a non-prototypical part of speech. This task is situated within a natural language inference paradigm. We test the abilities of five language models -- two proprietary models (GPT-3.5 and GPT-4), three open-source models (Mistral 7B, Falcon 40B, and Llama 2 70B). We find that GPT-4 performs best on the task, followed by GPT-3.5, but that the open source language models are also able to perform it and that the 7B parameter Mistral displays as little difference between its baseline performance on the natural language inference task and the non-prototypical syntactic category task, as the massive GPT-4.

Recent research in provably secure neural linguistic steganography has overlooked a crucial aspect: the sender must detokenize stegotexts to avoid raising suspicion from the eavesdropper. The segmentation ambiguity problem, which arises when using language models based on subwords, leads to occasional decoding failures in all neural language steganography implementations based on these models. Current solutions to this issue involve altering the probability distribution of candidate words, rendering them incompatible with provably secure steganography. We propose a novel secure disambiguation method named SyncPool, which effectively addresses the segmentation ambiguity problem. We group all tokens with prefix relationships in the candidate pool before the steganographic embedding algorithm runs to eliminate uncertainty among ambiguous tokens. To enable the receiver to synchronize the sampling process of the sender, a shared cryptographically-secure pseudorandom number generator (CSPRNG) is deployed to select a token from the ambiguity pool. SyncPool does not change the size of the candidate pool or the distribution of tokens and thus is applicable to provably secure language steganography methods. We provide theoretical proofs and experimentally demonstrate the applicability of our solution to various languages and models, showing its potential to significantly improve the reliability and security of neural linguistic steganography systems.

To reduce the inference cost of large language models, model compression is increasingly used to create smaller scalable models. However, little is known about their robustness to minority subgroups defined by the labels and attributes of a dataset. In this paper, we investigate the effects of 18 different compression methods and settings on the subgroup robustness of BERT language models. We show that worst-group performance does not depend on model size alone, but also on the compression method used. Additionally, we find that model compression does not always worsen the performance on minority subgroups. Altogether, our analysis serves to further research into the subgroup robustness of model compression.